Classification of The Human Development Index in Kalimantan Using Random Forest Method

Abstract

Random Forest is an ensemble classifier in a machine-learning algorithm. The ensemble classifier aimed to improve model accuracy and classification performance. Based on accuracy measures, Random Forest shows the best performance with existing ensemble classifiers like Support Vector Machine (SVM) and AdaBoost. Hence, this research will classify Human Development Index in Kalimantan using a Random Forest classifier. The predictor variables of classification are the average length of schooling, adjusted per capita additions, life expectancy, and length of school expectations. The Random Forest showed that the number of trees selected was 500 and the m being tried was 2. Adjusted per capita additions were the most influential variable in the increase of The Human Development Index with an importance of 10,61%. The Accuracy of classification was 58,33%.